• Journal of The Korean Society of Integrative Medicine
• /
• v.12 no.3
• /
• pp.11-24
• /
• 2024

Purpose : This study aims to identify the factors influencing health-related quality of life through machine learning of the general characteristics of patients with hypertension and to provide a basis for related research on patients, such as intervention strategies and management guidelines in the field of physical therapy for health promotion. Methods : Annual data from the second Korean Health Panel (Version 2.0) from 2019 to 2020, conducted jointly by the Korea Health and Social Research Institute and the National Health Insurance Service, were analyzed (Korea Health Panel, 2024). The data used in this study was collected from January to July 2020, and the data was collected using computer-assisted face-to-face interviews. Of the 13,530 household members surveyed, 1,368 were selected as the final study participants after removing missing values from 3,448 individuals diagnosed with hypertension by a doctor. Results : The results showed that walking (P2) was the most significant factor affecting health-related quality of life in random forest, followed by perceived stress (HS1), body mass index (BMIc), total household income (TOTc), subjective health status (SRHc), marital status (Marr), and education level (Edu). Conclusion :To prevent and manage chronic diseases such as hypertension, as well as to provide customized interventions for patients in advanced stages of the disease, research should be conducted in the field of physical therapy to identify influencing factors using machine learning. Based on the findings of this study, we believe that there is a need for additional content that can be utilized in the field of physical therapy to improve the health-related quality of life of patients with hypertension, such as diagnostic assessment and intervention management guidelines for hypertension, and education on perceived stress and subjective health status.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.113-120
• /
• 2024

This study aims to assess the effectiveness of machine learning models in predicting the pass rates of physical therapy students in national exams. Traditional grade prediction methods primarily rely on past academic performance or demographic data. However, this study employed machine learning and deep learning techniques to analyze mock test scores with the goal of improving prediction accuracy. Data from 1,242 students across five Korean universities were collected and preprocessed, followed by analysis using various models. Models, including those generated and fine-tuned with the assistance of ChatGPT-4, were applied to the dataset. The results showed that H2OAutoML (GBM2) performed the best with an accuracy of 98.4%, while TabNet, LightGBM, and RandomForest also demonstrated high performance. This study demonstrates the exceptional effectiveness of H2OAutoML (GBM2) in predicting national exam pass rates and suggests that these AI-assisted models can significantly contribute to medical education and policy.

• PNF and Movement
• /
• v.18 no.2
• /
• pp.287-296
• /
• 2020

Purpose: The study aims were to develop a wearable inertial sensor-based gait analysis device that uses machine learning algorithms, and to validate this novel device using temporal gait parameters. Methods: Thirty-four healthy young participants (22 male, 12 female, aged 25.76 years) with no musculoskeletal disorders were asked to walk at three different speeds. As they walked, data were simultaneously collected by a motion capture system and inertial measurement units (Reseed®). The data were sent to a machine learning algorithm adapted to the wearable inertial sensor-based gait analysis device. The validity of the newly developed instrument was assessed by comparing it to data from the motion capture system. Results: At normal speeds, intra-class correlation coefficients (ICC) for the temporal gait parameters were excellent (ICC [2, 1], 0.99~0.99), and coefficient of variation (CV) error values were insignificant for all gait parameters (0.31~1.08%). At slow speeds, ICCs for the temporal gait parameters were excellent (ICC [2, 1], 0.98~0.99), and CV error values were very small for all gait parameters (0.33~1.24%). At the fastest speeds, ICCs for temporal gait parameters were excellent (ICC [2, 1], 0.86~0.99) but less impressive than for the other speeds. CV error values were small for all gait parameters (0.17~5.58%). Conclusion: These results confirm that both the wearable inertial sensor-based gait analysis device and the machine learning algorithms have strong concurrent validity for temporal variables. On that basis, this novel wearable device is likely to prove useful for establishing temporal gait parameters while assessing gait.

• Journal of the East Asian Society of Dietary Life
• /
• v.18 no.1
• /
• pp.80-86
• /
• 2008

Both hand-made and machine-made Chinese noodles are popular in Korea. In this study, each type was evaluated in terms of its physical properties to rigorously determine for consumers which one has better qualities. The noodles were instrumentally measured for color, size, moisture content, density, viscoelasticity, and cutting force. The behaviors of the noodles were visually observed during cooking, and sensory evaluations were performed with the cooked noodles. The hand-made raw noodles were less dense, had higher moisture content, and generated more bubbles during cooking than the machine-made noodles. This indicated that the hand-made noodles contained more entrapped air, thereby resulting in the above physical characteristics. The change in noodle size after cooking was greater in the hand-made noodles, indicating that more entrapped air in expansion escaped during cooking and was replaced by water. The cutting force and viscoelasticity of the hand-made noodles were lower, and were controlled by viscous properties, respectively. These results agreed with the fact that the hand-made noodles had higher moisture content and lower density. In the sensory evaluation, the hand-made noodles presented lower hardness, but higher elasticity. It was inferred that the hand-made noodle dough underwent repeated processes of folding and extending, resulting in better developed of the gluten structure. Consequently, the hand-made noodles were determined to be different than the machine-made noodles in terms of instrumental measurements and sensory observations, suggesting that the hand-made noodles had superior textural properties.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.5
• /
• pp.1998-2014
• /
• 2016

In recent years, cloud computing services based on smart phones and other mobile terminals have been a rapid development. Cloud computing has the advantages of mass storage capacity and high-speed computing power, and it can meet the needs of different types of users, and under the background, mobile cloud computing (MCC) is now booming. In this paper, we have put forward a new classification-based virtual machine placement (CBVMP) algorithm for MCC, and it aims at improving the efficiency of virtual machine (VM) allocation and the disequilibrium utilization of underlying physical resources in large cloud data center. By simulation experiments based on CloudSim cloud platform, the experimental results show that the new algorithm can improve the efficiency of the VM placement and the utilization rate of underlying physical resources.

• Journal of the Korean Society of Physical Medicine
• /
• v.9 no.3
• /
• pp.293-299
• /
• 2014

PURPOSE: The purpose of this study was to investigate the effect of functional training using a sliding rehabilitation machine (SRM) on the mobility of the ankle joint and balance in children with cerebral palsy (CP). METHODS: The subjects consisted of 11 children who were diagnosed with spastic CP. They carried out the functional training using the SRM for 30 minutes, three times a week, for 8 weeks. Before and after all of the training sessions, the subjects were tested using the Pediatric Balance Scale (PBS) and Gross Motor Function Measurement (GMFM), range of motion (ROM) in the ankle joint, the pennation angle of the gastrocnemius muscle and the fascicle length of gastrocnemius muscle were measured to determine the mobility of the ankle joint and balance ability. RESULTS: There were significant differences between the pre-test and post-test in the PBS and GMFM. The ROM of the ankle joint was significantly increased after the functional training using the SRM. Moreover, the fascicle length was increased and the pennation angle was decreased after the functional training using the SRM, but the difference was not significant. CONCLUSION: These results suggest that functional training using the SRM may have some effect on the mobility of ankle joint and balance in children with CP. According to the results, this study could present an approach to the rehabilitation or treatment of children with CP.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1279-1290
• /
• 2021

As vehicle technology has grown, autonomous driving that does not require driver intervention has developed. Accordingly, CAN security, an network of in-vehicles, has also become important. CAN shows vulnerabilities in hacking attacks, and machine learning-based IDS is introduced to detect these attacks. However, despite its high accuracy, machine learning showed vulnerability against adversarial examples. In this paper, we propose a adversarial CAN frame generation method to avoid IDS by adding noise to feature and proceeding with feature selection and re-packet for physical attack of the vehicle. We check how well the adversarial CAN frame avoids IDS through experiments for each case that adversarial CAN frame generated by all feature modulation, modulation after feature selection, preprocessing after re-packet.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.8
• /
• pp.2801-2815
• /
• 2022

With the advent and realization of Software Defined Network (SDN) architecture, many organizations are now shifting towards this paradigm. SDN brings more control, higher scalability, and serene elasticity. The SDN spontaneously changes the network configuration according to the dynamic network requirements inside the constrained environments. Therefore, a monitoring system that can monitor the physical and virtual entities is needed to operate this type of network technology with high efficiency and proficiency. In this manuscript, we propose a real-time monitoring system for data collection and visualization that includes the Prometheus, node exporter, and Grafana. A node exporter is configured on the physical devices to collect the physical and virtual entities resources utilization logs. A real-time Prometheus database is configured to collect and store the data from all the exporters. Furthermore, the Grafana is affixed with Prometheus to visualize the current network status and device provisioning. A monitoring system is deployed on the physical infrastructure of the KOREN topology. Data collected by the monitoring system is further pre-processed and restructured into a dataset. A monitoring system is further enhanced by including machine learning techniques applied on the formatted datasets to identify the elephant flows. Additionally, a Random Forest is trained on our generated labeled datasets, and the classification models' performance are verified using accuracy metrics.

• Neurospine
• /
• v.15 no.4
• /
• pp.329-337
• /
• 2018

Objective: Machine learning algorithms excel at leveraging big data to identify complex patterns that can be used to aid in clinical decision-making. The objective of this study is to demonstrate the performance of machine learning models in predicting postoperative complications following anterior cervical discectomy and fusion (ACDF). Methods: Artificial neural network (ANN), logistic regression (LR), support vector machine (SVM), and random forest decision tree (RF) models were trained on a multicenter data set of patients undergoing ACDF to predict surgical complications based on readily available patient data. Following training, these models were compared to the predictive capability of American Society of Anesthesiologists (ASA) physical status classification. Results: A total of 20,879 patients were identified as having undergone ACDF. Following exclusion criteria, patients were divided into 14,615 patients for training and 6,264 for testing data sets. ANN and LR consistently outperformed ASA physical status classification in predicting every complication (p < 0.05). The ANN outperformed LR in predicting venous thromboembolism, wound complication, and mortality (p < 0.05). The SVM and RF models were no better than random chance at predicting any of the postoperative complications (p < 0.05). Conclusion: ANN and LR algorithms outperform ASA physical status classification for predicting individual postoperative complications. Additionally, neural networks have greater sensitivity than LR when predicting mortality and wound complications. With the growing size of medical data, the training of machine learning on these large datasets promises to improve risk prognostication, with the ability of continuously learning making them excellent tools in complex clinical scenarios.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.387-388
• /
• 2020

As the utilization of microservice architectural style in diverse applications are increasing, the microservice deployment cost became a concern for many companies. We propose an approach to reduce the deployment cost by generating an algorithm which minimizes the cost of basic operation of a physical machine and the cost of resources assigned to a physical machine. This algorithm will produce optimal resource allocation and deployment location based on genetic algorithm process.